Abstract title: “ Growing through innovation replicability ...

1  

Abstract No: 025-0834

Abstract title: “ Growing through innovation replicability: The role of investment in routine adaptation”

Juliana Hsuan*

Copenhagen Business School Department of Operations Management

Solbjerg Plads 3 DK-2000 Frederiksberg, Denmark

[email protected]

Moren Lévesque

Schulich School of Business York University

4700 Keele Street Toronto, ON, Canada M3J 1P3 [email protected]

POMS 23rd Annual Conference Chicago, Illinois, U.S.A.

April 20 to April 23, 2012   

* Corresponding author

2  

GROWING THROUGH INNOVATION REPLICABILITY: THE ROLE OF INVESTMENT IN ROUTINE ADAPTATION

We propose a formal model of firm growth through replication that considers the extent of the

investment to adapt routines as replication unfolds and the portion of this investment that goes

toward innovation in the routines. The use of these two investment constructs brings about four

types of growth policies. We use a utility function that considers proxies for both growth and

failure potential to uncover the role played in selecting these policies by the economic

environment of the targeted market for expansion. Our analysis further reveals the importance of

the innovation-relative-to-imitation investment efficiency in adapting the routines to be

replicated while selecting a growth policy. The refinement of a replication theory through our

analysis of growth policies result in two testable hypotheses.

Keywords: firm growth; replication; innovation; routine adaptation; utility maximization

1. Introduction

Most economies encourage firm growth, because it generates economic prosperity by creating

jobs (e.g., Minniti and Lévesque, 2008, 2010). A means by which firms can grow more rapidly is

through replication. Replication refers to “the creation and operation of a large number of similar

outlets that deliver a product or perform a service” (Winter and Szulanski, 2001, p.730). In

accordance with Winter’s (2010) distinctions, we consider in this article spatial replication,

whereby the growth arises via capacity expansion at a new location, rather than temporal

replication with adding capacity at the current location. Many successful examples of spatial

replication exist in the franchising industry (e.g., Jonsson and Foss, 2011).

3  

Szulanski and Jensen (2008) note that “[g]iven that organizational routines embody a

firm’s productive knowledge […] competitive advantage on a global scale may require the

replication of successful practices” (p.1733). These practices are short-run routines (e.g., method

of payments) that determine the firm’s operating characteristics, investment routines (e.g.,

employee training), or routines that modify over time various aspects of the operating

characteristics (e.g., product features to respond to competitive pressures) (Nelson and Winter,

1982). When the firm chooses to grow through (spatial or temporal) replication, these routines

(existing or new to the firm) will require some form of investment for adaptation (Aspara et al.,

2010; Rivkin, 2001). Consequently, what growth policy should a firm employ considering (1) the

requirement of investment while replicating to adapt its routines to the newly targeted market,

(2) the reduction of the threat of imitation through innovation in these routines, and (3) the

deployment of more replication effort for higher levels of investment or innovation? Addressing

this question enables us to put forward a typology that categorizes firm growth policies as bear-

or bull-innovation and bear- or bull-imitation.

We differentiate adaptation from innovation by considering the effort (in term of

investment) required to adapt the firm’s business routines contingent on the newness in these

routines. Newness arises because the routine is either new to the market or new to the firm. Take

the Swedish furniture company IKEA, for instance, which introduced in 1963 a new routine to

the Norwegian market with its product distribution system, where buying customers could easily

transport their fit-in-a-flat-box furniture. In the 60s, a more standard furniture distribution system

(in Norway or elsewhere) was home delivery. The routine is new to the firm when, instead, its

smallness (in the case of a new venture) implies that it still lacks many of the routines required to

run a firm of a larger size and it must allocate resources to either create new routines or find and

4  

implement the best routines of high-performing firms (Csaszar and Siggelkow, 2010). Imitation

rather than innovation takes place when the firm can copy in the targeted market the routines that

it has already in place in the source environment. Our treatment of innovation thus differs from

Szulanski and Jensen’s (2008, p.1733) “innovation in the form of local adaptation” (and is with

respect to firms’ routines rather than their product offerings1).

Current formulas for success do not indicate what degree of innovativeness (i.e.,

newness) should be embedded in the firm’s business routines as it attempts to grow. Nor do they

inform us on what the sensitivity of the effort allocated toward routine innovation should look

like as the targeted market environment improves or deteriorates in comparison to the source

environment. Yet, Zahra’s (2011) classification of business opportunity exploitation, as being

either ‘entrepreneurial’ or ‘routine,’ calls for the consideration of a continuum between these two

extremes based on the degree of innovation in the firm’s routines. We thus address the gap in the

literature by formulating a formal model that offers a complementary theory of firm growth that

is based on routine innovation and replication.

We focus on new business ventures (i.e., small new firms) and therefore consider their

three key features: lack of routines, lack of resources, and lack of reputation (e.g., Joglekar and

Lévesque, in press). The lack of routines required to grow through replication can arguably be an

advantage, since ‘nothing is set in stone,’ thus providing the firm with more flexibility (Sapienza

et al., 2006). The lack of resources, such as investment to adapt theirs or others’ routines to be

replicated, may force the ventures to be more efficient in using the little they have (e.g.,

Mosakowski, 2002). The lack of reputation, however, often leads to higher risk of failure (due to

liability of smallness and newness, or legitimacy; Singh et al., 1986) and limited growth

1 We refer the interested reader to Forsman (2011) for a discussion on innovation typologies.

5  

potential.2 By integrating these shortages in a decision framework, along with the extent of the

investment to adapt routines and the portion that goes toward innovation, we show that a new

venture’s growth policies must be guided by changes in the external economic environment it

will face, as well as changes in the efficiency of the innovation-relative-to-imitation investment

to adapt the routines to be replicated.

While also considering the knowledge transfer advantage through the relative amount of

effort allocated to the potentially hard-to-imitate routines (Rivkin, 2001), we thus simultaneously

bring two key dimensions to the forefront: investment decisions and environmental changes,

especially as they pertain to the venture’s expected return and risk from the newly targeted

market in comparison to the source environment. We use innovation effort as a mechanism that

limits or prevents imitation by influencing the firm’s level of replication effort. Our findings also

speak to the performance-imitation relationship (Csaszar and Sigglekow, 2010) by describing

when the venture should strive for a bear-imitation policy, where a small investment in routine

adaptation with a small portion for innovation (or relatively more imitation) is made (e.g., where

only a few known routines are required for the targeted market and most can be copied from the

routines the venture has already in place in the source environment). We also bring additional

light to the theory on replication and adaptation (e.g., Williams, 2007) by characterizing when to

strive for a bull-imitation policy, where a large investment in routine adaptation with a small

portion for innovation is made. We contribute to the work on the timing and scale of innovation

in replication (Szulanski and Jensen, 2006, 2008) by articulating conditions under which

suppressing innovation may not necessarily be a growth policy to follow for the venture.

2 Short et al. (2009) find empirical evidence to contest this view, arguing that the niche market positioning of many new ventures and today’s changing economy can relax the liability-of-newness concern for survival.

6  

In the next section we review seminal studies on replication to clarify key constructs and

better articulate our contributions. We then present a formal model that enables us to refine our

theoretical arguments, followed by an analysis of growth policies for new business ventures. The

proposed prescriptions offer a number of theoretical implications that we discuss prior to

concluding with two testable hypotheses, their managerial implications and future research.

2. Growth through Replication and Innovation

The objective of replication is “to expand the scope of the existing accomplishment, avoiding

‘reinventing the wheel’ ” (Winter, 2010, p.100). Replicators typically grow through repeated

application of a simple formula or recipe, as seen with global giants like McDonalds, Starbucks

or Walmart. Replication can also be viewed as a mechanism for exploring the extent to which

superior routines and distinctive capabilities contribute to the likelihood of firm growth and

survival (or failure). Superior routines and distinctive capabilities depend on processes that

involve the combination of physical resources and human collaboration as repositories for firm

tacit and explicit knowledge (Olavarrieta and Ellinger, 1997). As they define the heterogeneity of

resources and internal capabilities of the firm, these routines can determine the extent to which

the firm’s offering is inimitable by competitors (Dierickx and Cool, 1989).

The replication literature has mostly focused on the growth of multi-national firms

(Jonsson and Foss, 2011), franchises (e.g., Szulanski and Jensen, 2008), firms in the service

industries (e.g., Williams, 2007; Winter and Szulanski, 2001; Voss and Hsuan, 2009), and the

replication of business models (e.g., Aspara et al., 2010; Dunford et al., 2010; Teece, 2010). Of

particular interest is the study by Williams (2007) that treats replication as a mechanism for

knowledge transfer and “an effort aimed at creating activities at one location that are identical to

those at another location” (p.869). When knowledge is more self-contained (i.e., with less

7  

influence from external factors), firms do not have to consider external links, hence incurring

lower costs of, or effort for, replication. Winter and Szulanski (2001) also bring the concept of

knowledge transfer by introducing the ‘Arrow core’ to capture which attributes are replicable

and worth replicating, and how these attributes are created.

Aspara et al. (2010) look instead at replication via the exploitation of an innovative

business model that articulates “the logic and provides data and other evidence that demonstrates

how a business creates and delivers value to customers” (Teece, 2010, p.173). Aspara et al. find

that, unlike large firms, small ones exhibit strong profitable growth when they have a high

emphasis on business model innovation, yet a low emphasis on replicating this model. Our focus

on the growth of business ventures through the replication of routines requires us to pay a special

attention to knowledge transfer by means of replication effort deployment. It also requires paying

attention to the venture’s business model, especially as it affects the investment made in routine

adaptation and the portion that goes toward innovation.

The concepts of innovation and imitation in replication are therefore at the forefront of

our study. Winter and Szulanski (2001) articulate that “[r]eplicators are a subset of the class of

organizations whose founding and/or early growth is closely tied to a particular innovation”

(p.731). Rivkin (2001) points out that entrepreneurs’ goal is to have a superior system of

production that they can replicate themselves, but not be imitated by others. Replication can thus

be a firm’s competitive advantage (enabling it to grow) if imitation from rivals can be prevented.

That can be the case, while replication is made easier, when the firm possesses compatible

specifications and protocols within its boundaries or across industry. Rivkin (2001) further

identifies game-theoretic strategic maneuvers by incumbents as potential barriers to imitation,

but not to replication, since productive knowledge can be duplicated within a firm at no cost.

8  

This literature also adds that, within the firm, standardization creates flexibility and

makes it easier to replicate through learning and routines (Nelson and Winter, 1982), while

outside the firm, standardization facilitates imitation (Rivkin, 2001). Although these studies have

investigated some of the strategic growth implications of replication, the formulas for success do

not indicate the level of innovation that should be embedded in a venture’s routines as it attempts

to expand. Nor do they inform us on what the effort allocated to creating new routines, finding

and implementing the best routines of high-performing firms, or implementing the routines that

the venture masters but that are new to a market (i.e., effort toward innovation) should look like

as the venture’s targeted market differs from its source environment. Our approach can attend to

these strategic issues.

Additionally, investing in routine adaptation plays a key role in firm growth. Williams

(2007) considers replication (i.e., exact copying or cloning) and adaptation (or modification) as

two separate knowledge transfer mechanisms, suggesting that “[f]irms need to differentiate

between ambiguous knowledge, which must be copied exactly, and knowledge intertwined with

the environment, which requires modification” (p.884). Williams also observes that the use of

adaptation increases significantly with the time elapsed, or experience gained, in a partnering

relationship. Szulanski and Jensen (2008) argue instead that “[i]nnovation in the form of local

adaptation, if undertaken too early or too substantially may undermine the ability to use the

template, or original set of practices, thus hampering the replication” (p.1733).

Jonsson and Foss (2011) further note that the retail part of IKEA could not neglect the

heterogeneity of its targeted markets, because successful replication is highly dependent on

product range, pricing, standard operating procedures, etc., all of which needing to be adapted

locally. Due to small-scale economies drawn from market adaptation, rather than large-scale

9  

ones enabled by standardization (Kaufmann and Eroglu, 1998), this local adaptation requires the

firm to make irreversible and costly investments in order to support its many retail activities. The

model we now introduce considers the crucial role played by investment in routine adaptation, in

addition to being well-grounded in the above replication literature.

3. A Growth Model based on Innovation Replicability

We develop our arguments using a decision-theoretic model. Such model is desirable to refine

and build theory of decision-making, especially when considering the dynamics born from

simultaneous changes in factors that influence the outcome of the decision to be made. A

decision-theoretic model is also desirable when the objective is to deepen our understanding of

the tradeoffs that may be involved. In our context, these tradeoffs are not only guided by the

wealth that can be generated through replication, but also by the variability surrounding that

wealth, thus considering return (or growth) as well as risk (survival) potentials.

Sapienza et al. (2006) argue for the need to consider both firm growth and firm survival

as new firms attempt to initiate their internationalization. Such consideration is particularly

important since firm survival (without which a firm cannot grow) is far from guaranteed for any

new business ventures (e.g., Headd, 2003). Published numbers on business failure over the past

decade have been astonishingly scary. The statistics for the U.S. has been at 40% of new

ventures failing within two years and 80% within five (Small Business Administration, 2009).

Numbers are slightly more encouraging for Sweden, with Short et al.’s (2009) representative

sample from the late 90s showing a 48% rate of failure within the first five years. Deepening our

understanding of what might yield firm growth can only happen without disregarding what might

lead to firm failure. We use utility theory to build an objective function that considers both

10  

growth and failure, but first we describe model parameters and decision variable that enable us to

suggest a typology of growth policies.

3.1. Adaptation Investment, Innovation and Replication Efforts

Consider a new business venture attempting to grow by adding capacity, or replicate, at a newly

targeted location. The expansion depends on the venture’s ability to deploy, and thus adapt,

various business (or organizational) routines in the new market. We represent by i the investment

for routine adaptation for this expansion, and by the proportion of that investment that goes

toward innovation (i.e., creating new routines, finding and implementing the routines of best-

performing firms, or implementing the routines mastered by the venture that are new to the

targeted market). The proportion , which we also refer to as innovation effort, is a decision to be

made by the venture owner, while the routine adaptation investment i is considered to be given

based on the venture’s financial capabilities.3 These two constructs are attractive because they

are tangible and measurable (also facilitating empirical measurement and testing); that is, the

venture owner is likely able to identify their values.4

Replication is not straightforward (Winter, 2010); we represent by e the replication effort.

This effort depends on both, the amount of routine adaptation investment i as well as the

innovation effort . Williams (2007) argues that, due to context dependence on knowledge

transfer leading to replication, not only copying but also adaptation (modification) is required.

Rivkin (2001) further argues that what eases replication also eases imitation, thus suggesting that

hard-to-imitate innovation diminishes the ease of replicability. Hence, we put forward a positive

relationship between e and i and between e and . That is, larger investment or greater 3 The symmetry of i and in the replication effort function we define results in similar technical findings (although with a change of interpretation) if instead is given and i becomes the decision variable. 4 Forsman (2011) warns that innovation development in small firms may not always be easily distinguishable, since the corresponding activities can be viewed as “quality improvements in collaboration with customers” (p.740).

11  

innovation is expected to require more effort to replicate the venture’s routines in order to grow.

Arguably, more investment requires more learning and changes to properly use the routines (in

the existing or new market), with a greater portion of that investment toward innovation further

accentuating the learning and changes.

We swipe a large number of possible (multiplicative) functional forms for these

relationships by using a power (or Cobb-Douglas type) function. Formally, , where

(> 0) is the productivity (in terms of the required replication effort) of the investment, and

exponents a and b, both positive, represent the elasticity of the innovation effort and of the

routine adaptation investment, respectively.5 Taken together these three parameters characterize

the efficiency of replication effort (i.e., the efficiency of the innovation-relative-to-imitation

investment to adapt the routines to be replicated), where less replication effort is required per

investment dollar for small values of , b and a than for large values. This functional form is also

attractive for empirical testing of the relationship between our two investment constructs and the

replication effort, as its logarithmic transformation results in a linear (e.g., regression) function.

We further characterize (albeit now qualitatively) the relationship between these three

key constructs by providing a typology of new business ventures. We label the venture as being a

‘bear-imitator’ in its source environment when both, routine adaptation investment i and

innovation effort , are low (and hence e is low). An online delivery of fresh produces could fit

this profile. Its expansion would likely involve some investment in adapting its goods

sourcing/delivery, selection, and web design, but it would not be substantial, nor would be a need

for creating new routines or finding/implementing the routines of best-performing firms. Most

likely, this venture would deploy the routines it already masters and that are known to the

5 Without a loss of generality, we could have also written the functional form as , where i represents the innovation investment.

12  

targeted market, thus imitating what is already in place to expand. This characterization

somewhat differs from Winter’s (2010), which views the imitator acting more independently in

that “the knowledge transfer occurs without the active support of the source” (p.102).

When i is high but is low, there is a significant amount of routine adaptation but not

much innovation effort. We thus employ the term ‘bull-imitator’ for labeling the venture in its

source environment. A distributor of generic drugs could fit this profile, because it would likely

encounter substantial investment (to expand) while adapting its distribution channels (e.g.,

hospitals, pharmacies, general practitioners, etc.) as well as order sizes and product selection for

customization, just to name a few. However, the need for creating new routines and

implementing the routines of best-performing firms and those it masters but that are new to the

targeted market (i.e., its required innovation effort for expansion) could be negligible when

using, for instance, existing distribution channels and a mix-and-match of the generic drugs.

When both i and are high, we employ the term ‘bull-innovator,’ since the actual amount

of investment toward routine innovation (i.e., i) would be substantial. We could consider Voss

and Hsuan’s (2009) small mortgage bank to exemplify this type. The key routines included IT

design and compatibility across branches, employee training for call centers, accessibility to

services through multiple channels, and customization of service packages. The bank had to

substantially invest in routine adaption to compete, as it required the alignment of the front office

(e.g., customization options for customers) with the back office activities (e.g., IT development

to ensure multi-channel access). In addition, the portion of its routine adaptation investment

toward innovation was substantial, because it had to create new routines (e.g., IT design) and

bring those that it mastered but that were new to the targeted market (e.g., service packages).

13  

We finally coin the term ‘bear-innovator’ for when i is low but is high; that is, when

money is scarce yet innovation must take place. A venture that created face recognition software

could fit this profile. It could have been working with hardware manufacturers to sell the

software as part of a bundle, where a computer recognizes who the person is and uses this in

several areas (e.g., another means of identity for authentication, personalizing the experience,

etc.). The venture could have decided to try and sell the same exact software to law enforcement

agencies, since the software could also be used to find a person in a crowd (or to notice if a

particular person went by a camera). Although the software is the same, which keeps routine

adaptation investment low, the routines on how to sell to this new niche market are different,

thus the need to deploy more innovation efforts to create new routines.

The resulting typology is summarized in Table 1(a). We return to this table (parts a and

b) once we analyze the dynamics of innovation replicability to argue how a new business venture

might be best advised to move from one growth policy to another (since the environment of the

targeted market for expansion is unlikely to be identical to the source environment). However,

we must first describe the venture’s utility function.

------------------ (Insert Table 1 about here) ------------------ 3.2. Optimizing Utility

The venture accumulates wealth, denoted W, which can encompass its sales or sales growth,

regarded as the most important measures of new venture performance (Robinson and

McDougall, 2001), or even assets or other key items that are considered for valuation purposes.

The wealth is inversely related to the replication effort e. Indeed, as more effort must be

allocated toward replication, less effort remains available for other crucial venture activities that

can generate wealth, which is especially an issue for small firms with limited resources (Aspara

14  

et al., 2010; Ebben and Johnson, 2005). The wealth is also affected by some random external

effects (e.g., those emerging from the competitive landscape, from new governmental

regulations, or from industrial regulation changes) that escape the control of the venture. We

aggregate these effects into a random variable X (viewed as a random percentage of change in

the venture’s wealth due to effects outside its control) that is normally distributed with mean X

and standard deviation X. Factors under the control of the venture (e.g., timing of entry or

supplier selection) further affect the wealth. We again aggregate these firm-based factors into a

parameter, , since our primarily goal is to tease out the impact of replication effort e on the

wealth, not these factors’ impacts. Formally, or, using the functional form for e,

.6

The venture owner wishes to maximize utility from generating wealth, where the more

the wealth the more the utility. However, the venture owner has a certain attitude toward risk in

that he/she is risk-averse. Caliendo et al. (2009) report that 79% of self-employed individuals

exhibited low-to-medium willingness to take risks, while Elston and Audretsch (2011) found that

the majority of entrepreneurs they sampled were risk-averse. We thus represent the utility by an

exponential function U with , where is positive to characterize the owner’s

risk aversion.7 The use of this exponential utility function, along with the normally distributed

external effects X (and thus W given its linear relationship with X), makes the venture owner

indifferent, for any alternative W, between maximizing expected utility and selecting the

maximum of the certainty equivalent E(W) – ½Var(W) (e.g., Freund, 1956). E() is the

6 Without a loss of generality, the exponent of e can be 1 (because , a and b can be rescaled accordingly). 7 The underlying assumption is that the decision maker possesses a constant absolute risk aversion (or CARA). The use of other utility functions (e.g., constant relative risk aversion [CRRA] or utility or prospect theory) is at the cost of losing mathematical tractability.

15  

expectation operator, Var() the variance operator, and the certainty equivalent is a risk-adjusted

wealth. Given the amount of investment i toward routine adaptation, the venture must select the

proportion of that investment that goes toward routine innovation so as to maximize expected

utility; i.e., * that maximizes .

We highlight a few facts about this certainty equivalent. We begin with the first term, the

expected wealth; that is, a proxy for the likelihood of firm growth. This growth proxy decreases

(whenever 1 + X > 08) as more replication effort ( ) is required. Indeed, easing

replication, thus requiring less effort, enables the venture to gain from learning effects, develop

abilities to serve more markets, or perhaps shape industry structures, which in turn facilitate the

generation of more wealth and enable growth. Also, more effort remains available for other

crucial activities of the resource-limited venture (Aspara et al., 2010; Ebben and Johnson, 2005).

Furthermore, the level of investment i in routine adaptation moderates this negative relationship

in that more investment makes the relationship less negative (i.e., ), because there is

more of a (financial) cushion to address or absorb the negative consequences of deploying more

replication effort.

As for the second term of the certainty equivalent, it represents a proxy for the likelihood

of firm failure and also decreases as more replication effort is required. This negative

relationship between failure and deployed replication effort is expected, because easing

replication exposes the venture to imitation and erodes its competitive advantages (e.g., Porter,

1996). The level of investment i in routine adaptation also moderates the negative relationship in

that more investment makes the relationship less negative (i.e., ), because this

8 In other words, the expected percentage of change in the venture’s wealth due to external effects must not be worse than –100%, which is expected for the business to be going in the first place.

16  

additional (financial) cushion can limit the variability in lost wealth resulting from more

replication effort. Figure 1 illustrates these moderating effects as well as the tradeoffs between

the growth proxy (return) and the failure proxy (risk).

------------------ (Insert Figure 1 about here) ------------------

4. Analysis: Growth Policy Dynamics

The decision model just described enables us to identify the innovation effort (i.e., the portion of

routine adaptation investment puts toward innovation) that maximizes the venture owner’s

utility. We investigate how a change in a model parameter affects this effort. We then move

away from a static view and look at how the simultaneous evolutions of key model parameters

during expansion influence the trajectory of innovation efforts. This exercise enables us to bring

back the typology in Table 1 to investigate how the dynamics of innovation replicability might

move a venture from one growth policy to another.

We straightforwardly find the unique optimal innovation effort to be .9

This * corresponds to the level where additional utility from expected wealth resulting from a

change in this effort equals the additional disutility associated with the variation in the wealth (a

marginal-revenue-equals-marginal-loss argument). We can characterize eight key relationships

from this derivation. The first five are negative relationships between * and the venture’s

adaptation investment i, the productivity of that investment (in terms of replication effort), the

elasticity a of the innovation effort, the elasticity b of the routine adaptation investment, and the

mean percentage X of the change in wealth due to external effects. The next three are positive

9 This * is formally derived by taking the first-order derivative of the certainty equivalent with respect to . Note

also that the second-order derivative with respect to is and we

straightforwardly verify that once is replacing in that equation then it becomes negative. In other

words, the certainty equivalent is concave at *, which makes * the unique optimal solution.

17  

relationships between * and the marginal wealth from firm-based factors (aggregated), the

standard deviation X of the change in wealth due to external effects, and the venture owner’s

risk attitude . The utility-disutility tradeoff born from an increase in any of these parameters

depends on how easing growth through replication by restricting innovation outweighs the

potential damage from imitation. Table 2 summarizes this sensitivity analysis on *.

------------------ (Insert Table 2 about here) ------------------

The characterization of these negative and positive relationships enables us to find

conditions under which the venture should decrease or increase its innovation effort based on

simultaneous changes in the routine adaptation investment, in the efficiency of replication effort,

and in the economic environment that it will face for the newly targeted market. In taking this

more dynamic view, we first note that risk attitude α is a stable personality trait not expected to

change across different situations that may arise over time (e.g., Weber and Milliman 1997). We

can thus reasonably assume that it is fixed. We can also keep fixed the marginal wealth from

aggregated firm-based factors, because the venture has the option of altering its inversely

proportional relationship between wealth and replication effort via the efficiency of this effort

when investment are made to adapt routines (without a loss of generality). We recall that this

efficiency is controlled by the adaptation investment’s productivity and elasticity b, as well as

by the elasticity a of the innovation effort.

Therefore, taken together, , b and a characterize whether replication effort efficiency

improves (when at least one of them decreases, the others staying the same) or deteriorates

(when at least one of them increases, the others staying the same) over time. This efficiency

brings about the first condition in our analysis of the dynamics of growth policies. The second

condition characterizes changes in the economic environment of the newly targeted market as

18  

compared to the venture’s source environment. Specifically, it involves the mean percentage X

of the change in wealth due to external factors (in the new market) and the standard deviation X

of this change. The economic environment is improving when the venture experiences an

increase in X but no increase (i.e., stays as is or decreases) in X, since the new environment will

be more rewarding and no more, or even less, risky. It is however deteriorating when the venture

experiences a decrease in X but no decrease (i.e., stays as is or increases) in X. We next use the

four resulting scenarios to describe utility-maximizing policies for growth (i.e., routine

adaptation investment i and its portion for innovation ). While, for each scenario, Table 3 sums

up the impacts of simultaneous changes in model parameters on the innovation effort, Table 4

summarizes the growth policies and Table 5 illustrates their movements for expansion.

------------------ (Insert Tables 3, 4 and 5 about here) ------------------

Scenario 1: The venture faces improvement in both the new environment and the

replication effort efficiency. Our decision framework (Table 2) proposes that the venture

increases the routine adaptation investment (i) sufficiently to counteract the impact on innovation

effort (*) (an increase) from the improved replication effort efficiency (, a and/or b decrease).

As a result, the improved environment (X increases and/or X decreases) and high investment in

routine adaptation enable the venture to decrease its (relative) innovation effort, that is, the

portion of this investment that goes toward routine innovation. We therefore suggest a bull-

imitation policy (with i high and low) for the targeted market, as portrayed in Table 1(b). From

this policy and the venture typology in Table 1(a), a bull-imitator in the source environment

should remain a bull-imitator (keep i high and low) for the newly targeted market, but all

others―bear-imitators, bear- and bull-innovators―should become bull-imitators. But if the

venture cannot afford a sufficient increase in the routine adaptation investment, our framework

19  

instead proposes that it decreases this investment sufficiently to counteract the impact on the

innovation effort (a decrease) from the improved economic environment. Yet, the improved

replication effort efficiency and lower investment in routine adaptation enable the venture to

increase the portion of this investment that goes toward innovation. We thus suggest a bear-

innovation policy (with i low and high) for the targeted market, and a bear-innovator should

remain so, while all others―bear- and bull-imitators, as well as bull-innovators―should become

bear-innovators.

Scenario 2: The venture faces improvement in the new environment but deterioration in

the replication effort efficiency. The deteriorated replication effort efficiency encourages the

deployment of less innovation effort (because greater innovation effort also means greater

replication effort). Hence, our framework proposes that the venture either keeps as is or increases

the investment in routine adaptation (given the improved environment), but decreases the portion

of this investment that goes toward innovation (i.e., relative innovation effort). We therefore

suggest a bull-imitation policy for the targeted market, leaving some room for the bear-imitators

(with both i and low). Specifically, from this policy and the typology offered in Table 1(a), a

bear-imitator in the source environment should stay as is or become a bull-imitator, a bull-

imitator should stay as is, a bear-innovator should become a bear- or a bull-imitator, and a bull-

innovator should become a bull-imitator.

Scenario 3: The venture faces deterioration in the new environment but improvement in

the replication effort efficiency. The deteriorated environment encourages innovation, but not so

much investment in routine adaptation. Our framework proposes that the venture either keeps as

is or decreases the routine adaptation investment (given the deteriorated environment), but

increases the portion of this investment that goes toward innovation (i.e., relative innovation

20  

effort). We therefore suggest a bear-innovation policy for the targeted market, leaving some

room for the bull-innovators (with both i and high). Specifically, a bear-imitator should

become a bear-innovator, a bull-imitator should become a bull- or a bear-innovator, a bear-

innovator should stay as is, and a bull-innovator should stay as is or become a bear-innovator.

Scenario 4: The venture faces deterioration in both the new environment and the

replication effort efficiency. This last scenario requires similar growth policies to that of Scenario

1, albeit for different reasons. Our decision framework proposes that the venture increases the

investment in routine adaptation sufficiently to counteract the impact on innovation effort (an

increase) from the deteriorated environment. Nevertheless, both the deteriorated environment

and replication effort efficiency encourage the venture to decrease its (relative) innovation effort

(since more innovation yields more replication effort), that is, the portion of the investment that

goes toward routine innovation. We thus suggest a bull-imitation policy for the newly targeted

market, where a bull-imitator should remain so, while all others―bear-imitators, bear- and bull-

innovators―should become bull-imitators. But when the venture cannot afford a sufficient

increase in the routine adaptation investment, our framework proposes that it decreases this

investment sufficiently to counteract the impact on the innovation effort (a decrease) from the

deteriorated replication effort efficiency. The deteriorated environment and lower investment

encourage the venture to increase the portion of this investment that goes toward innovation to

gain some competitive advantages. We thus suggest a bear-innovation policy, where a bear-

innovator should stay as is, while all others should become bear-innovators.

We can exemplify some of these scenarios by considering whether the venture

contemplates growth in an emerging as opposed to a developed market. Scenario 2 might best

represent growth in an emerging market, where the efficiency of replication effort is expected to

21  

deteriorate since the structures in the business environment (e.g., intellectual property

regulations, product safety laws, etc.) that are in place are likely less stringent than in the source

environment. But the economic environment is likely more vibrant than the source environment,

and the venture can likely attract more funding to adapt its routines because financers are eager

to explore these new markets. In this context, the growth policy should leave some room for

bear-imitation for those who are already bear-imitators or bear-innovators in their source

environment, but mostly focus on bull-innovation for the newly targeted market.

Scenario 3, on the other hand, might best represent firm growth in developed markets that

experience deteriorating economic conditions, as seen in some of today’s power economies of

the West. The efficiency of replication effort is expected to increase from known and established

structures in the business environment, as well as from the venture’s experiences if the source

environment is already in a developed market. However, developed (new) markets may not

enable the venture to attract more investment, because it may already be financially saturated (as

part of the deteriorating economic conditions). In this context, the growth policy for the newly

targeted market should mostly focuses on bear-innovation, but also leave some room for bull-

innovation for those who are already bull-innovators or bull-imitators in their source

environment.

5. Discussion

These prescriptions yield a number of theoretical and practical implications. Winter and

Szulanski’s (2001) approach to replication is grounded in the economics of information and

entails a broad scope of knowledge transfer as well as dynamic capabilities through routinized

activities. “The resource leveraged by true replication strategy is an informational resource, and

information is not ‘spread thin’ by being used at multiple locations” (p.734). They further argue

22  

that the replication strategy is resistant to imitation (i.e., sustainable), because replicators have

better access to key routines, learn from experiences, possess firm-level capabilities unavailable

to others, have an organizational history that enables superior transfers of knowledge, and a tight

fit exists between these knowledge-based advantages and activities. We use a different angle by

grounding our arguments in the economics of innovation.

Specifically, we use innovation effort (i.e., the portion of investment in routine adaptation

that goes toward innovation by, e.g., creating new routines, finding and implementing the best

routines of high-performing firms, or implementing the routines mastered by the venture that are

new to a market) as the mechanism that limits or prevents imitation by influencing the firm’s

level of replication effort. This approach enables us to characterize conditions that are based on

the efficiency of replication effort and the economic environment of the newly targeted market,

under which a variety of growth policies might be suboptimal to a utility-maximizing venture

owner. For instance, investing more in routine adaptation and increasing the portion of that

investment that goes toward innovation, which conditions characterize a bull-innovation policy,

should be beneficial only if the venture’s targeted new market is associated with improved

replication effort efficiency, but a deteriorated economic environment.

In our approach to firm growth through replicability, we also view the knowledge transfer

advantage through the relative amount of effort allocated to routine innovation. This view echoes

Rivkin’s (2001) in that “knowledge that is moderately complex lends itself to being replicated

without being imitated” (p.287), where complexity is measured by the number of decisions a

firm must make and of interactions among these decisions. Greater innovation effort thus likely

leads to greater complexity. Rivkin’s explanation for ‘replication-without-imitation’ lies with the

managerial search for knowledge, where the imitator has to discover while the replicator, instead,

23  

only has to rediscover. This rediscovery corresponds to an easier transfer of knowledge as

compared to the imitator’s discovery, since the replicator enjoys the benefits (e.g., learning) of its

original template. Our framework, albeit not focused on the replication-imitation connection,

brought two key dimensions omitted by Rivkin (2001) to the forefront: investment decisions and

environmental changes, both affecting the search for knowledge. New business ventures are

especially sensitive to these two dimensions (e.g., formal investors may control where their

invested money goes), and we do find that they affect growth policies.

Our framework also leaves room for the imitator. Csaszar and Sigglekow (2010) put

forward conditions under which greater investment in imitating the practices of high-performing

firms can be valuable. These conditions involve firm and context similarity (e.g., same industry

maturity), complexity (based on interactions among practices), and the timeframe during which

the benefit of imitation is observed. In the short-run, their findings differ from common wisdom

by proposing a negative relationship between firm performance and low levels of imitation

breath (i.e., expected number of successfully copied practices) when context similarity and

complexity are both high. A negative relationship also stands almost consistently in the short-run

when context similarity is low. Although our analysis compares in the short-run the

environments that the venture faces―rather than the environment of the firm it copies―to its

own, our findings also speak to the performance-imitation relationship. New ventures should

strive to grow via a bear-imitation policy (i.e., low levels of investment in routine adaptation

with small portions toward innovation) when the targeted market’s economic environment

improves as compared to the source environment, but the venture cannot afford to change its

routine adaptation investment and in fact should make the innovation portion even smaller (i.e.,

24  

mostly copies in the targeted market the routines that it has already in place in the source

environment) because the efficiency in replication effort is going to deteriorate.

Moreover, our operationalization of adaptation differs from Williams (2007) who views

replication (exact copying) and adaptation (modification) as separate mechanisms of knowledge

transfer. For us, adaptation is through investments rather than based on a combined scaling for

the use of, among other things, modified practices from partners. Nevertheless, we bring

additional light to Williams’ claims in that knowledge transfer is context dependent and a change

in the economic environment can put the role of adaptation and, as a result, bull-imitation

policies to the forefront. We propose that firms should strive to use such growth policies (of

large routine adaptation investments with small portions for innovation) when the targeted

market’s economic environment improves, as compared to that of the source environment, and

the venture can afford a sufficient investment increase. But even when the new environment is

worse, a venture should still favor adaptation, and the corresponding bull-imitation policy, if

efficiency in the replication effort is also going to deteriorate yet it can afford a sufficient

increase in the routine adaptation investment.10

Lastly, Szulanski and Jensen (2006, 2008) highlight the harm that can be caused by

undertaking innovation too substantially or too soon as the firm tries to grow. Staying more

faithful to the original template can be more favorable than moving away from it through

innovation, especially in the early stage of firm growth. Szulanski and Jensen propose a negative

relationship between innovation and growth in the early stage of growth, but a positive one (or at

least diminishing returns from imitating) in the later stage when the firm has gained more

10  The domination of the bull-imitation policy we observed might also be connected to the ‘principle-based’ as opposed to ‘template-based’ replication strategy put forward by Baden-Fuller and Winter (2007) (see also Winter, 2010). While following the template encourages imitation (or close copying) and to follow people ‘who already know how to do it,’ following principles, instead, encourages ‘presumptive’ adaptation and only use guidance from people ‘who already know how to do it’ to reach means-ends.

25  

experience. Our findings may have implications for these views, albeit for a different setting,

where suppressing innovation may not necessarily be the venture’s policy to follow. Indeed, a

bull-innovation policy (i.e., large investments in routine adaptation with large portions for

innovation) should be beneficial when the venture’s targeted new market is associated with

improved replication effort efficiency, but a deteriorated economic environment. On the other

hand, cash-constrained ventures should strive to implement a bear-innovation policy (i.e., small

investments in routine adaptation with large portions for innovation) when the targeted market’s

economic environment deteriorates as compared to the source environment and the venture

cannot afford to increase this investment, perhaps even having to decrease it. But even when the

new environment gets better, business ventures should still favor a bear-innovation policy when

efficiency in the replication effort is going to also be better yet the venture is forced to

significantly decrease its routine adaptation investment.

6. Conclusion and Extensions

Our formal model enabled the identification of tradeoffs and helped us refine our theoretical

arguments. Key to this model was the extent of the investment to adapt routines, considering the

portion of this investment that went toward innovation, used to develop a typology of growth

policies. Solving for a certainty equivalent (utility maximization) that considers proxies for both

growth and failure potential, uncovered the crucial role played by (1) changes in the external

economic environment of the targeted market as compared to the source environment, and (2)

the replication effort efficiency or, in other words, changes in the efficiency of the innovation-

relative-to-imitation investment to adapt the routines to be replicated in this new market. The

refinement of a replication theory through our analysis of firm growth policies and their

implications with respect to existing literature can be summarized by two testable hypotheses.

26  

These hypotheses give a central role to the bull-imitation and bear-innovation policies, as also

evoked in Table 5:

HYPOTHESIS 1. When the economic environment of the newly targeted market improves over

the source environment, the venture will generally favor a bull-imitation policy for its

expansion, with one exception:

if the venture cannot afford a greater investment in routine adaptation for the new

market, even though its replication effort efficiency would be better in this new market,

then a bear-innovation policy will be favored.

HYPOTHESIS 2. When the economic environment of the newly targeted market deteriorates as

compared to the source environment, the venture will generally favor a bear-innovation

policy for its expansion, with one exception:

if the venture can afford a greater investment in routine adaptation for the new market,

even though its replication effort efficiency would be worse in this new market, then a

bull-imitation policy will be favored.

These hypotheses also have implications for the management of established firms. With

their fatter financial cushions, established firms are more likely to be able to afford a greater

investment in routine adaptation for the new market than new business ventures can.

Consequently, established firms would be best advised to adopt a bull-imitation policy and stick

to it as they move from one new market to the next, even when the economic environment of the

targeted market deteriorates as compared to the source environment. Under this policy, the

established firm would copy in the targeted market its numerous routines already in place in the

source environment. However, the large routine adaptation investment under such a policy, even

with a relatively small portion for routine innovation, could still result in a considerable amount

of new routines being created and the routines of best-performing firms, or those that the

27  

established firm masters but new to a market, being implemented. That is, routine innovation

could still take place in a much larger extent than it could in a financially-constrained venture.

Nevertheless, financially-constrained ventures could be less so with, for instance, a well-

developed venture-capital market, as is the case in the U.S., U.K. and Israel for various

technology sectors. Although a number of European countries (e.g., Italy) have smaller stock

exchanges, Bertoni et al. (2011) show that venture-capital financing has been positively

associated with the growth of new technology ventures in Italy. Bertoni et al. also suggest that,

as European policy-makers attempt to increase the demand for venture capital (e.g., by

increasing the birth rate of technology firms with subsidies) as well as its supply (e.g., through

co-investment schemes or government ownership of such funds), more technology ventures are

likely to emerge. Our hypotheses propose that the growth of these emerging firms rely on

adopting a bull-imitation policy under which substantial routine innovation can be afforded.

In addition to empirically testing the above two hypotheses, multiple avenues for

extensions to overcome this work’s limitations exist. First, a simulation approach could be used

to extend our theoretical argument by involving tradeoffs born from the competitive landscape

over time. The simulation would include multiple firms (incumbents and new entrants)

interacting in an industry where each would grow or die. Each firm could be endowed with a

certain amount of investment for routine adaptation and a portion for innovation (perhaps

randomly determined as it enters the landscape) and select its growth policy according to Table

1’s typology. We could then evaluate how these policies fair with respect to each other, or else

they could be randomly assigned to the set of firms. We could then also evaluate how new

entrants fair with respect to their more established counterparts. The firm owner’s decision could

28  

again be based on a risk-adjusted wealth to capture a proxy for growth and failure potential, or

both could be studied separately.

A second extension could make the investment in routine adaptation and the portion for

innovation (or even the outcomes from this investment, i.e., replication effort) random. de Jong

and Marsili (2006) maintain that, of roughly a third of small firms that have a formal plan for

innovation, only half put aside a budget for innovation. Added to their lack of resources, this

observation suggests that most new ventures may not have a ‘budget for replication’ either

(although replication through franchising may overcome the resource scarcity problem; Combs

et al., 2011). Instead, their owners must search for funding, not knowing whether they will be

successful in generating it and/or generating enough; von Burg and Kenny (2000) noted that it

took nearly two years for Network Systems, the first venture to offer local area networking, to

receive venture capital funds. Furthermore, more investments (especially toward innovation) do

not necessarily guarantee immediate better outcomes, because payoffs may lag in time. These

new dimensions―randomness and delay―to refine our decision framework could also be

addressed with simulation. The use of simulation has already produced some significant insights

in the replication literature (see, e.g., Rivkin, 2001).

Although replication may not be the sole mechanism that can produce large-scale

economic value from an original innovation, Winter (2010) contends that invention without

innovation is not creating economic value, but innovation without replication is not creating

much economic value. Our treatment of innovation has been on innovation in a firm’s routines

(i.e., by creating new routines, finding and implementing the best routines of high-performing

firms, or implementing the routines mastered by the venture that are new to a market) rather than

on its products or services. Nevertheless, such innovation is crucial in the context of replication

29  

for new business ventures that typically lack already developed routines. New ventures also lack

resources that can ease the need for innovation replicability, making it more likely an important

competitive advantage. A third impediment common in new ventures is the lack of reputation or

‘transferable reputation’ given the need for “real uniformity in the customer experience” as

replication takes place (Winter, 2010, p.101). While paying attention to these shortages, much is

left to learn about the growth of new business ventures and innovation replicability.

Acknowledgements

We are grateful to Chris Voss and the participants of the 6 October 2011 Institut for Produktion

og Erhvervsøkonomi’s Research Seminar at Copenhagen Business School for their insightful

suggestions on earlier versions of this manuscript. We also acknowledge financial support from a

travel grant from the Institut for Produktion og Erhvervsøkonomi for a guest professorship at the

Copenhagen Business School, a Standard Research Grant (#410-2011-1004) from the Social

Sciences and Humanities Research Council (SSHRC) of Canada, and a Discovery Grant

(#341679-2007) from the Natural Sciences and Engineering Research Council (NSERC) of

Canada.

References

Aspara, J., Hietanen, J., Tikkanen, H., 2010. Business model innovation vs replication: Financial

performance implications of strategic emphases. Journal of Strategic Marketing 18(1), 39-56.

Baden-Fuller, C., Winter, S.G., 2007. Replicating organizational knowledge: Principles or templates?

Wharton School, University of Pennsylvania working paper.

Bertoni, F., Colombo, M.G., Grilli, L., 2011. Venture capital financing and the growth of high-tech

startups: Disentangling treatment from selection effects. Research Policy 40, 1028-1043.

30  

Caliendo, M., Fossen, F.M., Kritikos, A.S., 2009. Risk attitudes of nascent entrepreneurs: New evidence

from an experimentally validated survey. Small Business Economics 32, 153-167.

Combs, J.G., Ketchen, D.J. Jr, Short, J.C., 2011. Franchising research: Major milestones, new directions,

and its future within entrepreneurship. Entrepreneurship Theory and Practice 35, 413-425.

Csaszar, F.A., Sigglekow, N., 2010. How much to copy? Determinants of effective imitation breadth.

Organization Science 21(3), 661-676.

de Jong, J.P.J., Marsili, O., 2006. The fruit flies of innovations: A taxonomy of innovative small firms.

Research Policy 35(2), 213-229.

Dierickx, I., Cool, K., 1989. Asset stock accumulation and sustainability of competitive advantage.

Management Science 35(12), 1504-1511.

Dunford, R., Palmer, I., Benveniste, J., 2010. Business model replication for early and rapid

internationalization: The ING direct experience. Long Range Planning 43, 655-674.

Ebben, J.J., Johnson, A.C., 2005. Efficiency, flexibility, or both? Evidence linking strategy to

performance in small firms. Strategic Management Journal 26, 1249-1259.

Elston, J.A., Audretsch, D.B., 2011. Financing the entrepreneurial decision: An empirical approach using

experimental data on risk attitudes. Small Business Economics 36(2), 209-222.

Forsman, H., 2011. Innovation capacity and innovation development in small enterprises: A comparison

between the manufacturing and service sectors. Research Policy 40, 739-750.

Freund, R., 1956. The introduction of risk into a programming model. Econometrica 24, 253-263.

Headd, B., 2003. Redefining business success: Distinguishing between closure and failure. Small

Business Economics 21(1), 51-61.

Joglekar, N., Lévesque, M., In press. The role of operations management across the entrepreneurial value

chain. Production and Operations Management.

Jonsson, A., Foss, N.J., 2011. International expansion through flexible replication: Learning from the

internationalization experience of IKEA. Journal of International Business Studies, available online.

31  

Kaufmann, P.J., Eroglu, S., 1998. Standardization and adaptation in business format franchising. Journal

of Business Venturing 14, 69-85.

Minniti, M., Lévesque, M., 2010. Entrepreneurial types and economic growth. Journal of Business

Venturing 25(3), 305-314.

Minniti, M., Lévesque, M., 2008. Recent developments in the economics of entrepreneurship. Journal of

Business Venturing 23(6), 603-612.

Mosakowski, E., 2002. Overcoming resource disadvantages in entrepreneurial firms: When less is more,

in Hitt, M., Ireland, D., Sexton, D., Camp, M. (Eds), Strategic Entrepreneurship: Creating an

Integrated Mindset. Blackwell, Oxford, U.K., pp.106-126.

Nelson, R.R., Winter, S.G., 1982. An Evolutionary Theory of Economic Change. Belknap Press,

Cambridge, Mass.

Olavarrieta, S., Ellinger, A.E., 1997. Resource-based theory and strategic logistics research. International

Journal of Physical Distribution & Logistics Management 27(9/10), 559-587.

Porter, M.E., 1996. What is strategy? Harvard Business Review 74(6), 61-78.

Rivkin, J.W., 2001. Reproducing knowledge: Replication without imitation at moderate complexity.

Organization Science 12(3), 274-293.

Robinson K.C., McDougall, P.P., 2001. Entry barriers and new venture performance: a comparison of

universal and contingency approaches. Strategic Management Journal 22(6-7), 659-685.

Sapienza, H.J., Autio, E., George, G. Zahra, S.A., 2006. A Capabilities perspective on the effect of early

internationalization on firm survival and growth. Academy of Management Review 31(4), 914-933.

Short, J.C., McKelvie, A., Ketchen, D.J., Chandler, G.N., 2009. Firm and industry effects on firm

performance: A generalization and extension for new ventures. Strategic Entrepreneurship Journal 3,

47-65.

Singh, J.V., Tucker, D.J., House, R.J., 1986. Organizational legitimacy and the liability of newness.

Administrative Science Quarterly 31, 171-193.

Small Business Administration, 2009. www.sba.gov/advo/stats/sbfaq.pdf.

32  

Szulanski, G., Jensen, R.J., 2008. Growing through copying: The negative consequences of innovation on

franchise network growth. Research Policy 37, 1732-1741.

Szulanski, G., Jensen, R.J., 2006. Presumptive adaptation and the effectiveness of knowledge transfer.

Strategic Management Journal 27, 937-957.

Teece, D.J., 2010. Business models, business strategy and innovation. Long Range Planning 43, 172-194.

von Burg, U., Kenney, M., 2000. Venture capital and the birth of the local area networking industry.

Research Policy 29, 1135-1155.

Voss, C.A., Hsuan, J., 2009. Service architecture and modularity. Decision Sciences 40(3), 541-569.

Weber, E.U., Milliman, R.A., 1997. Perceived risk attitudes: Relating risk perception to risky choice.

Management Science 43(2), 123-144.

Williams, C. 2007. Transfer in context: Replication and adaptation in knowledge transfer relationships.

Strategic Management Journal, 28, 867-889.

Winter, S.G. 2010. The replication perspective on productive knowledge, in: Itami, H., Kusonoki, K.,

Iumagami, T. (Eds.), Dynamics of Knowledge, Corporate Systems and Innovation. Springer-Verlag,

Berlin Heidelberg, pp. 95-121.

Winter, S.G., Szulanski, G. 2001. Replication as strategy. Organization Science, 12(6), 730-743.

Zahra, S. 2011. Opportunity exploitation: The core of strategic entrepreneurship. Keynote presentation at

the conference on Multi- and Micro-Level Issue in Strategic Entrepreneurship, Copenhagen Business

School, 14 October 2011.

33  

Figure 1. A replication framework

Table 1. Venture and growth policy typologies

(a) Labeling of ventures (b) Labeling of growth policies

Routine adaptation investment, i Routine adaptation investment, i Low High Low High

Innovation effort,

High Bear-innovator Bull-innovator Bear-innovation Bull-innovation Low Bear-imitator Bull-imitator Bear-imitation Bull-imitation

Table 2. Sensitivity of the (optimal) innovation effort

An increase in Notation Innovation effort * Reasons Routine adaptation investment i

Increases the replication effort, thus diminishing the utility from a lower expected wealth as well as the disutility from the variation in this wealth, which tradeoff can be offset by allocating less innovation effort to bring back down the replication effort

Productivity of the routine adaptation investment

Elasticity of the innovation effort

a

Elasticity of the routine adaptation investment

b

Mean percentage of change in wealth due to external effects X

Diminishes only the utility from a higher expected wealth, which can be offset by allocating less innovation effort and thus bringing back up this utility

Marginal wealth from firm-based factors (aggregated)

Augments the utility from a higher expected wealth as well as the disutility from the variation in this wealth (and at a faster rate), which tradeoff can be offset by allocating more innovation effort

Standard deviation of the change in wealth due to external effects

X Augments only the disutility from the variation in the wealth, which can be offset by allocating more innovation effort and thus bring back down this disutility Venture owner’s risk attitude

Legend: represents a decrease in *; represents an increase in * 

Expected wealth, E[W] (likelihood-of-firm-growth

proxy) –

increase (i.e., become less negative) + % of routine adaptation investment in innovation,

Replication effort, e

Routine adaptation investment, i

increase (i.e., become less negative)

Risk adjustment,  Var[W] (likelihood-of-firm-failure

proxy)

–

34  

Table 3. Impacts on the (optimal) innovation effort from simultaneous parameter changes

Scenario 1 The venture faces

improvement in both the new environment and the

replication effort efficiency

Scenario 2 The venture faces

improvement in the new environment but deterioration

in the replication effort efficiency

Scenario 3 The venture faces

deterioration in the new environment but

improvement in the replication effort efficiency

Scenario 4 The venture faces

deterioration in both the new environment and the

replication effort efficiency

Parameter

change Impact on * Parameter

change Impact on * Parameter

change Impact on * Parameter

change Impact on *

i a) enough b) enough

― or ― or ― or ― or a) enough b) enough

a

b

X

― ― ― ― ― ― ― ―

X

― ― ― ― ― ― ― ―

Legend: represents a decrease; represents an increase; ― represents no change

35  

Table 4. The dynamics of growth policies

Investment policy (i,) Economics of the new environment

Improving (X increases and/or X decreases) Deteriorating (X decreases and/or X increases)

Replication effort efficiency

Improving (, a and/or b decrease)

Scenario 1 a) Increase the routine adaptation investment (i) sufficiently (to counteract the impact of the improved efficiency in replication effort1) but decrease the portion of that investment that goes toward innovation () b) Decrease the routine adaptation investment (i) sufficiently (to counteract the impact of the improved external economic environment2) but increase the portion of that investment that goes toward innovation ()

Scenario 3 Do not increase (i.e., stay as is or decrease) the routine adaptation investment (i) but increase the portion of that investment that goes toward innovation ()

Deteriorating (, a and/or b increase)

Scenario 2 Do not decrease (i.e., stay as is or increase) the routine adaptation investment (i) but decrease the portion of that investment that goes toward innovation ()

Scenario 4 a) Increase the routine adaptation investment (i) sufficiently (to counteract the impact of the deteriorated external economic environment3) but decrease the portion of that investment that goes toward innovation () b) Decrease the routine adaptation investment (i) sufficiently (to counteract the impact of the deteriorated efficiency in replication effort4) but increase the portion of that investment that goes toward innovation ()

1 Since , , and are all negative, a sufficient increase in i will counteract the positive impact on * from a decrease in , a and b. 2 Since and are negative but is positive, a sufficient decrease in i will counteract the negative impact on * from an increase in and a decrease in . 3 Since and are negative but is positive, a sufficient increase in i will counteract the positive impact on * from a decrease in and an increase in . 4 Since , , and are all negative, a sufficient decrease in i will counteract the negative impact on * from an increase in , a and b.

36  

Table 5. Illustrating the dynamics of growth policies

Investment policy (i,) Economics of the new environment

Improving (X increases and/or X decreases) Deteriorating (X decreases and/or X increases)

Replication effort efficiency

Improving (, a and/or b decrease)

Scenario 1 a) can increase i b) cannot increase i

Scenario 3

Deteriorating (, a and/or b increase)

Scenario 2

Scenario 4 a) can increase i b) cannot increase i

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Bear-innovators

Bear-imitators

Bull-innovators

Bull-imitators

high

low

low high

Innovation efforts (ρ)

Routine adaption investment (i)